SRC Forum - Message Replies

Forum: Reliability & Maintainability Questions and Answers

Topic: Reliability & Maintainability Questions and Answers

Topic Posted by: Reliability & Maintainability Forum ( )
Organization: System Reliability Center
Date Posted: Mon Aug 31 12:47:36 US/Eastern 1998

Back to message list Show all replies Topics List About this forum
Original Message:

Posted by: Chua Peck Seng ( )
Organization:DSO National Laboratories
Date posted: Wed Apr 5 23:26:23 US/Eastern 2000
Subject: Reliability Test
May I know what are the reliability test to verify that a module can withstand intermittment operation at 50 degC for a life cycle of 8 yrs ?


Subject: Reliability Testing
Reply Posted by: Bruce Dudley ( )
Organization: Reliability Analysis Center
Date Posted: Wed Apr 12 15:45:31 US/Eastern 2000
There are two general types of tests that could be considered in demonstrating a 70,000 plus hour need for a module. The first is a qualification test at the highest operating temperature (50 degrees C) for at least 120,000 hours with no failures resulting in a 80% confidence. Multiple units could be tested to reduce the calendar test interval. The other test type is an accelerated stress test. This test requires the analysis of the module materials to determine the failure mechanisms and modes needed to accelerate the failure conditions. Care is needed in the application of the stress level to be sure that "overstressing" is not performed. Depending on the type of acceleration, test times could be reduced from the 100,000 plus hours to a lower level such as 10,000 to 20,000 hours. Some of the accelerated models that could be considered are: the Coffin-Manson Power Law for Mechanical Fatigue Damage, the Arrhenius Steady State Temperature Acceleration for Chemical Aging, The Peck Model for Combined Temperature and Humidity Effects, the Kemeny Model for Accelerated Voltage Testing and others. Developing valid and productive accelerated tests usually requires the need for technical specialist as there are no "magic" models that fit all conditions. Most "modules" have many material types and therefore many failure mechanisms. Determining the module wearout behavior requires special knowledge and experience as several competing behaviors may be present. Once the predominate failure conditions are determine, test models can be constructed to provide leverage in demonstrating an increased rate of information about the product life.

Reply to this message